Statistics in Public Mental Health: Our World Is Not Flat

• Activity Number: 112
• Type: Topic Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section for Statistical Programmers and Analysts
• Author(s): Alden Gross*+
• Companies: Institute for Aging Research

Abstract

A number of challenges underlie research in observational epidemiology, and in mental health research in particular. We rely on talking to a subject or on subjective measures of behaviors because there are almost no objectively measurable biologic processes for most mental health conditions. Study volunteers are not necessarily representative of a target population of real interest. Additionally, because of the nature of mental health disorders, their prevalence and characteristics vary by diagnostic fashions that change over time, how a question is asked, and how the sample is selected. Thus, a very important analytic challenge is in modeling mental disorders as latent constructs. Other common analytic and design challenges include multilevel influences and effects, statistical generalization in experimental and non-experimental settings, rare outcomes, need for longitudinal follow-up, and missing data. These exciting challenges require sound expertise in statistical methods. Some of the most influential researchers in public mental health during the past century have been statisticians, and their contributions are encouraged in this field.